Generalized score matching for non-negative data

Publikation: Bidrag til tidsskrift › Tidsskriftartikel › Forskning › fagfællebedømt

Standard

Generalized score matching for non-negative data. / Yu, Shiqing; Drton, Mathias; Shojaie, Ali.

I: Journal of Machine Learning Research, Bind 20, (76), 2019.

Publikation: Bidrag til tidsskrift › Tidsskriftartikel › Forskning › fagfællebedømt

Harvard

Yu, S, Drton, M & Shojaie, A 2019, 'Generalized score matching for non-negative data', Journal of Machine Learning Research, bind 20, (76).

APA

Yu, S., Drton, M., & Shojaie, A. (2019). Generalized score matching for non-negative data. Journal of Machine Learning Research, 20, [(76)].

Vancouver

Yu S, Drton M, Shojaie A. Generalized score matching for non-negative data. Journal of Machine Learning Research. 2019;20. (76).

Author

Yu, Shiqing ; Drton, Mathias ; Shojaie, Ali. / Generalized score matching for non-negative data. I: Journal of Machine Learning Research. 2019 ; Bind 20.

Bibtex

@article{c5de7446b11d46da868b5614bfda2777,
title = "Generalized score matching for non-negative data",
abstract = "A common challenge in estimating parameters of probability density functions is the intractability of the normalizing constant. While in such cases maximum likelihood estimation may be implemented using numerical integration, the approach becomes computationally intensive. The score matching method of Hyv{\"a}rinen (2005) avoids direct calculation of the normalizing constant and yields closed-form estimates for exponential families of continuous distributions over Rm. Hyv{\"a}rinen (2007) extended the approach to distributions supported on the non-negative orthant, Rm+ . In this paper, we give a generalized form of score matching for non-negative data that improves estimation efficiency. As an example, we consider a general class of pairwise interaction models. Addressing an overlooked inexistence problem, we generalize the regularized score matching method of Lin et al. (2016) and improve its theoretical guarantees for non-negative Gaussian graphical models.",
keywords = "Exponential family, Graphical model, Positive data, Score matching, Sparsity",
author = "Shiqing Yu and Mathias Drton and Ali Shojaie",
year = "2019",
language = "English",
volume = "20",
journal = "Journal of Machine Learning Research",
issn = "1533-7928",
publisher = "MIT Press",

}

RIS

TY - JOUR

T1 - Generalized score matching for non-negative data

AU - Yu, Shiqing

AU - Drton, Mathias

AU - Shojaie, Ali

PY - 2019

Y1 - 2019

N2 - A common challenge in estimating parameters of probability density functions is the intractability of the normalizing constant. While in such cases maximum likelihood estimation may be implemented using numerical integration, the approach becomes computationally intensive. The score matching method of Hyvärinen (2005) avoids direct calculation of the normalizing constant and yields closed-form estimates for exponential families of continuous distributions over Rm. Hyvärinen (2007) extended the approach to distributions supported on the non-negative orthant, Rm+ . In this paper, we give a generalized form of score matching for non-negative data that improves estimation efficiency. As an example, we consider a general class of pairwise interaction models. Addressing an overlooked inexistence problem, we generalize the regularized score matching method of Lin et al. (2016) and improve its theoretical guarantees for non-negative Gaussian graphical models.

AB - A common challenge in estimating parameters of probability density functions is the intractability of the normalizing constant. While in such cases maximum likelihood estimation may be implemented using numerical integration, the approach becomes computationally intensive. The score matching method of Hyvärinen (2005) avoids direct calculation of the normalizing constant and yields closed-form estimates for exponential families of continuous distributions over Rm. Hyvärinen (2007) extended the approach to distributions supported on the non-negative orthant, Rm+ . In this paper, we give a generalized form of score matching for non-negative data that improves estimation efficiency. As an example, we consider a general class of pairwise interaction models. Addressing an overlooked inexistence problem, we generalize the regularized score matching method of Lin et al. (2016) and improve its theoretical guarantees for non-negative Gaussian graphical models.

KW - Exponential family

KW - Graphical model

KW - Positive data

KW - Score matching

KW - Sparsity

UR - http://www.scopus.com/inward/record.url?scp=85072407015&partnerID=8YFLogxK

M3 - Journal article

AN - SCOPUS:85072407015

VL - 20

JO - Journal of Machine Learning Research

JF - Journal of Machine Learning Research

SN - 1533-7928

M1 - (76)

ER -

ID: 230391441